nuget-trusted-publishing
Set up NuGet trusted publishing (OIDC) on a GitHub Actions repo — replaces long-lived API keys with short-lived tokens. USE FOR: trusted publishing, NuGet OIDC, keyless NuGet publish, migrate from NuGet API key, NuGet/login, secure NuGet publishing. DO NOT USE FOR: publishing to private feeds or Azure Artifacts (OIDC is nuget.org only). INVOKES: shell (powershell or bash), edit, create, ask_user for guided repo setup.
- License 许可
- MIT
- 最近更新
- 2026-11-14
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- nuget-trusted-publishing
- license
- MIT
- description
- Set up NuGet trusted publishing (OIDC) on a GitHub Actions repo — replaces long-lived API keys with short-lived tokens. USE FOR: trusted publishing, NuGet OIDC, keyless NuGet publish, migrate from NuGet API key, NuGet/login, secure NuGet publishing. DO NOT USE FOR: publishing to private feeds or Azure Artifacts (OIDC is nuget.org only). INVOKES: shell (powershell or bash), edit, create, ask_user for guided repo setup.
02 资源文件 2
- 参考文档references/package-types.md8.2 KB
- 参考文档references/publish-workflow.md3.8 KB
03 SKILL.md 正文
NuGet Trusted Publishing Setup
Set up NuGet trusted publishing on a GitHub Actions repo. Replaces long-lived API keys with OIDC-based short-lived tokens — no secrets to rotate or leak.
Prerequisites
- GitHub Actions — this skill covers GitHub Actions setup only
- nuget.org account — the user needs access to create trusted publishing policies
When to Use This Skill
Use this skill when:
- Setting up trusted publishing for a NuGet package
- Migrating from
secrets.NUGET_API_KEYto OIDC-based publishing - Asked about keyless or secure NuGet publishing
- Creating a new NuGet publish workflow from scratch
- Asked to "remove NuGet API key" or "use NuGet/login"
- Setting up publishing for a dotnet tool, MCP server, or template package
- Asked about
NuGet/login@v1orid-token: write
Safety Rules
⚠️ Bail-out rule: If any phase fails after one fix attempt on an infrastructure/auth issue, stop and ask the user. Don't loop on environment problems.
⚠️ Never delete or overwrite without confirmation: Removing API key secrets, deleting tags/releases, removing workflow steps, or changing package IDs. NuGet package IDs are permanent — mistakes can't be undone.
Process
Fast-path for greenfield repos: When the user has a simple setup (one packable project, no existing publish workflow), don't gate on multi-turn assessment. Combine phases: create the workflow immediately, include nuget.org policy guidance, local pack recommendation, and filename-matching warning all in one response. The full phased process below is for complex or migration scenarios.
Phase 1: Assess
Inspect the repo and report findings before making any changes.
-
Find and classify packable projects — check
.csprojfiles andDirectory.Build.props(package metadata is often set repo-wide). Classify in this order (earlier matches win):<PackageType>Template</PackageType>→ Template<PackageType>McpServer</PackageType>→ MCP server (also a dotnet tool)<PackAsTool>true</PackAsTool>→ Dotnet tool- Class library (
IsPackable=trueor noOutputType) → Library <OutputType>Exe</OutputType>with<IsPackable>true</IsPackable>→ Application package (not a tool, but still publishable)<OutputType>Exe</OutputType>withoutPackAsToolorIsPackable→ Not packable by default (ask user if they intend to publish it)
-
Validate structure for each project's type:
Type Required All PackageId,Version(in .csproj or Directory.Build.props)Dotnet tool PackAsTool(required);ToolCommandName(optional but recommended — defaults to assembly name)MCP server PackageType=McpServer,.mcp/server.jsonincluded in packageTemplate PackageType=Template,.template.config/template.jsonunder content dir -
Find existing publish workflows in
.github/workflows/— look fordotnet nuget push,nuget push, ordotnet pack. -
Check version consistency — for MCP servers, verify
.csproj<Version>matches bothserver.jsonversion fields (rootversionandpackages[].version). Flag any mismatch. -
Report findings to the user: classification, missing properties, version mismatches, existing workflows. For multi-project repos, note whether one workflow or separate workflows per package are needed. Offer to fix gaps — use
ask_userbefore modifying project files.
❌ See references/package-types.md for per-type details and required properties.
Phase 2: Local Verification
Pack and verify locally before touching nuget.org — publishing errors waste a permanent version number.
⚠️ Always mention this step, even if you defer running it. Tell the user: "Before your first publish, run
dotnet pack -c Release -o ./artifactsto verify the .nupkg is created correctly."
dotnet pack -c Release -o ./artifacts— verify.nupkgis created- For tools/MCP servers: install from
./artifacts, run--help, uninstall - For libraries: inspect the
.nupkgcontents (it's a zip)
Phase 3: nuget.org Policy
This phase requires the user to act on nuget.org — guide them with exact values.
-
Determine the repo owner, repo name, and the workflow filename that will publish.
❌ The policy requires the exact workflow filename (e.g.,
publish.ymlorpublish.yaml) — just the filename, no path prefix. Matching is case-insensitive. Don't use the workflowname:field. -
Guide the user to create the trusted publishing policy:
Go to nuget.org/account/trustedpublishing → Add policy
- Repository Owner:
{owner} - Repository:
{repo} - Workflow File:
{filename}.yml - Environment:
release(only if the workflow usesenvironment:; leave blank otherwise)
Policy ownership: the user chooses individual account or organization. Org-owned policies apply to all packages owned by that org.
For private repos: policy is "temporarily active" for 7 days — becomes permanent after the first successful publish.
- Repository Owner:
-
Guide the user to create a GitHub Environment (recommended but optional — provides secret scoping + approval gates):
Repo Settings → Environments → New environment →
releaseAdd environment secret: Name =
NUGET_USER, Value = nuget.org username (NOT email)Optional: add Required reviewers for an approval gate.
⚠️ Wait for the user to confirm they've created the policy before asking them to remove old API keys/secrets or before attempting to run/publish with the workflow. Drafting or showing the workflow file itself is OK before confirmation.
Phase 4: Workflow Setup
Create or modify the publish workflow. The workflow must always be created or shown in your response — you may draft/show it even if the nuget.org policy is not yet confirmed, but do not guide the user to actually run/publish or remove old secrets until after confirmation.
Greenfield: Create publish.yml from the template in references/publish-workflow.md. Adapt .NET version, project path, and environment name. Ensure your output explicitly mentions id-token: write and NuGet/login@v1.
Migration (existing workflow with API key): Modify in place —
-
Add OIDC permission and environment to the publishing job:
jobs: publish: environment: release permissions: id-token: write # Required — without this, NuGet/login fails with 403 contents: read # Explicit — setting permissions overrides defaults -
Add the NuGet login step before push:
- name: NuGet login (OIDC) id: login uses: NuGet/login@v1 with: user: ${{ secrets.NUGET_USER }} # nuget.org profile name, NOT email -
Replace the API key in the push step:
--api-key ${{ steps.login.outputs.NUGET_API_KEY }} --skip-duplicate -
Verify: Ask the user to trigger a publish and confirm the package appears on nuget.org.
❌ Don't delete the old API key secret until trusted publishing is verified. Removing it is a one-way door — wait for confirmation.
Troubleshooting
| Problem | Cause | Fix |
|---|---|---|
NuGet/login 403 |
Missing id-token: write |
Add to job permissions |
| "no matching policy" | Workflow filename mismatch | Verify exact filename on nuget.org |
| Push unauthorized | Package not owned by policy account | Check policy owner on nuget.org |
| Token expired | Login step >1hr before push | Move NuGet/login closer to push |
| "temporarily active" policy | Private repo, first publish pending | Publish within 7 days |
already_exists on push |
Re-running same version | Add --skip-duplicate |
| GitHub Release 422 | Duplicate release for tag | Delete conflicting release (confirm first) |
| Re-run uses wrong YAML | gh run rerun replays original commit's YAML |
Delete obstacle, re-run — never re-tag |
⚠️ If any blocker persists after one fix attempt, stop and ask the user.
References
- Package type details: references/package-types.md — detection logic, required properties, minimal .csproj examples
- Publish workflow template: references/publish-workflow.md — complete tag-triggered workflow ready to adapt
- Microsoft docs: NuGet Trusted Publishing
04 同仓库更多 Skills
rust-best-practices
Guide for writing idiomatic Rust code based on Apollo GraphQL's best practices handbook. Use this skill when: (1) writing new Rust code or functions, (2) reviewing or refactoring existing Rust code, (3) deciding between borrowing vs cloning or ownership patterns, (4) implementing error handling with Result types, (5) optimizing Rust code for performance, (6) writing tests or documentation for Rust projects.
postgres-database-migration
Use this skill for planning, testing, and safely executing PostgreSQL schema migrations — especially when working with production data or shared databases. **Trigger when user asks to:** - Test a schema migration before applying it to production - Add, remove, or rename columns safely on a live table - Change a column's data type without downtime - Add or drop indexes, constraints, or foreign keys on large tables - Understand which ALTER TABLE operations lock the table - Roll back a failed migration - Plan a zero-downtime migration strategy - Fork a database to test a migration safely Covers: lock-level reference for every common DDL operation, safe migration patterns, fork-based testing, zero-downtime column changes, index creation, constraint addition, backfill strategies, pre/post-migration validation, and rollback planning.
postgres-best-practices
Best practices and guidelines for working with Postgres. Covers schema design, indexing strategies, query optimization, migrations, and common pitfalls. Use when writing SQL, designing database schemas, optimizing queries, or setting up a Postgres database.
writing-mstest-tests
ALWAYS USE when asked to fix, rewrite, update, improve, modernize, show corrected code for, or explain existing MSTest tests or MSTest-specific configuration. Use for "review" when corrected code or edits are wanted, even for one pasted assertion or passing tests with bad failure output. Covers expected/actual labels; generic Boolean, collection, string, numeric, null, identity, exception, hard-cast, and object[] checks; TestContext/lifecycle; timeout/cancellation; OS/CI conditions, retry, cleanup, parallelization, MSTest.Sdk project setup, and MSTESTxxxx. Honor the installed MSTest version. DO NOT USE to design new test cases (code-testing), perform report-only audits, create project files rather than explain MSTest setup, run tests, migrate frameworks, or handle non-MSTest/non-.NET code.
winforms-expert
Create, modify, debug, or review Windows Forms applications only when the request contains a concrete Windows Forms marker. USE FOR: WinForms, Windows Forms, System.Windows.Forms, WinForms Form or UserControl designer files (*.Designer.cs or *.Designer.vb), Visual Studio WinForms Designer, TableLayoutPanel, BindingSource, DataGridView, Control.InvokeAsync, component-tray ownership, or custom control serialization. DO NOT USE FOR: WPF or WPF XAML, including Window.InputBindings, KeyBinding, commands, or DataContext; .NET MAUI; Avalonia; or any request where none of the listed Windows Forms markers is present.
vectorization
Design, implement, optimize, and review SIMD code in .NET. USE FOR: vectorizing scalar loops with TensorPrimitives, Vector64/128/256/512, or platform hardware intrinsics; reviewing existing SIMD code, including the generic Vector type, for contract equivalence, tail handling, memory safety, portability, fallbacks, and measured performance. DO NOT USE FOR: performance work unrelated to SIMD or vectorization.